Aspect-based sentiment analysisA scalable system, a condition miner, and an evaluation dataset.

Résumé

Aspect-based sentiment analysis systems are a kind of text-mining systems that specialise in summarising the sentiment that a collection of reviews convey regarding some aspects of an item. There are many cases in which users write their reviews using conditional sentences; in such cases, mining the conditions so that they can be analysed is very important to improve the interpretation of the corresponding sentiment summaries. Unfortunately, current commercial systems or research systems neglect conditions; current frameworks and toolkits do not provide any components to mine them; furthermore, the proposals in the literature are insufficient because they are based on hand-crafted patterns that fall short regarding recall or machine learning models that are tightly bound with a specific language and require too much configuration. In this dissertation, we introduce Torii, which is an aspect-based sentiment analysis system whose most salient feature is that it can mine conditions; we also introduce Kami, which provides two deep learning proposals to mine conditions; and we also present Norito, which is the first publicly available dataset of conditions. Our experimental results prove our proposals to mine conditions are similar to the state of the art in terms of precision, but improve recall enough to beat them in terms of F1 score. Finally, it is worth mentioning that this dissertation would not have been possible without the collaboration of Opileak, which backs up the industrial applicability of our work.